Synaptic weight noise during multilayer perceptron training: fault tolerance and training improvements
Synaptic weight noise during multilayer perceptron training: fault tolerance and training improvements
复制标题
多层感知器训练期间的突触权重噪声：容错和训练改进
DOI：
10.1109/72.238328
复制
发表时间：
1993
影响因子：
--
通讯作者：
P. J. Edwards
中科院分区：
文献类型：
--
作者：
A. Murray;P. J. Edwards
建立了突触算术噪声对多层感知器训练影响的数学模型。预测了容错和泛化能力的增强以及学习轨迹的改进。这些预测随后通过仿真得到了验证。这些结果是非常普遍的，并对多层感知器(MLP)训练的精度要求有深刻的影响，特别是在模拟领域。
The authors develop a mathematical model of the effects of synaptic arithmetic noise in multilayer perceptron training. Predictions are made regarding enhanced fault-tolerance and generalization ability and improved learning trajectory. These predictions are subsequently verified by simulation. The results are perfectly general and have profound implications for the accuracy requirements in multilayer perceptron (MLP) training, particularly in the analog domain.